Aerial oxidation of tetrahydrofuran to 2-hydroxotetrahydrofuran in the presence of a trimeric CuI complex [Cu3L3] (HL = tBuNHC(S)NHP(S)(OiPr)2) and trapping of the unstable product at recrystallization

Article abstract of DOI:10.1039/c0nj00314j

The reaction of the potassium salt of N-thiophosphorylated thiourea tBuNHC(S)NHP(S)(OiPr)₂ (HL) with Cu(NO₃)₂ in aqueous EtOH leads to the trinuclear [Cu₃(tBuNHC(S)NP(S)(OiPr) ₂-S,S′)₃] ([Cu₃L₃]) complex. It was established that [Cu₃L₃] provokes the aerobic oxidation of tetrahydrofuran to 2-hydroxotetrahydrofuran and traps the latter at crystallization. The Royal Society of Chemistry and the Centre National de la Recherche Scientifique.

Full text of DOI:10.1039/c0nj00314j

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/njc | New Journal of Chemistry
Aerial oxidation of tetrahydrofuran to 2-hydroxotetrahydrofuran
in the presence of a trimeric CuI complex [Cu3L3]
(HL = tBuNHC(S)NHP(S)(OiPr)2) and trapping
of the unstable product at recrystallizationw
Robert C. Luckay,a Xia Sheng,a Christoph E. Strasser,a
Helgard G. Raubenheimer,a Damir A. Safin,*b Maria G. Babashkinab
and Axel Klein*b
Received (in Montpellier, France) 26th April 2010, Accepted 1st July 2010
DOI: 10.1039/c0nj00314j
The reaction of the potassium salt of N-thiophosphorylated thiourea tBuNHC(S)NHP(S)(OiPr)2
(HL) with Cu(NO3)2 in aqueous EtOH leads to the trinuclear [Cu3(tBuNHC(S)NP(S)(OiPr)2-S,S0)3]
([Cu3L3]) complex. It was established that [Cu3L3] provokes the aerobic oxidation of
tetrahydrofuran to 2-hydroxotetrahydrofuran and traps the latter at crystallization.
the structure of such compounds, as a rule, is impossible
Introduction
without the use of single crystal X-ray diffraction techniques.
Polynuclear CuI complexes are of great interest due to their
On the other hand, THF is important in biology. Several
luminescent properties,1 their use as catalysts,2 as precursors
actinomycetes, including Rhodococcus ruber strain 219,11
for chalcogenide nanoparticles3 and as models for biological
Pseudonocardia dioxanivorans,12 P. tetrahydrofuranoxydans
systems.4 There is a growing family of ligands based on
strain K1,13 and Pseudonocardia strains M114 and ENV47815
dithio(seleno)phosphinic acids RR0P(X)YH (X, Y = S, Se)
can utilize THF as the sole source of carbon and energy under
and imidodithio(seleno)diphosphinate R2P(X)NHP(X)R02
aerobic conditions. The pathway of THF degradation by these
(IDP), forming polynuclear aggregates with CuI.3a,5 In
organisms involves an initial mono-oxygenation of THF to
contrast to the previously mentioned ligands, there is a lack
generate the unstable hemiacetal, 2-hydroxytetrahydrofuran
of information about the structures of polynuclear CuI
(THF-2-ol). This intermediate is then rapidly dehydrogenated
complexes containing N-thiophosphorylated thioureas and
thioamides, RC(S)NHP(S)R02 (R = R02N, alkyl, aryl), which
to form 4-butyrolactone, a cyclic ester that subsequently
undergoes an esterase-catalyzed hydrolytic cleavage to form
are asymmetrical analogues of IDPs, containing a thiocarbonyl
4-hydroxybutyrate (Scheme 1).
group instead of one of the thiophosphinic units. The mole-
This hydroxyacid is further oxidized to succinate via
cular structures of only five polynuclear CuI complexes,
succinate semialdehyde, which enters the tricarboxylic acid
namely the cyclic trimers [Cu3{Et2NC(S)NP(S)(OPh)2}3],6
cycle. Another potential product of the abiotic dissociation of
[Cu3{morpholyn-N-ylC(S)NP(S)(OiPr)2}3] ([Cu3QI3]),7 the
2-hydroxytetrahydrofuran, 4-hydroxybutyraldehyde, may
hexanuclear aggregate [(Cu3{PhNHC(S)NP(S)(OiPr)2}3)2]
also be involved in THF degradation by R. ruber strain 219
([(Cu3QII3)2]),8 the supramolecular ‘‘honeycomb’’ aggregate
but it is more likely to be an intermediate in the catabolism of
1,4-butanediol, another precursor to 4-hydroxybutyrate.11
Based on gene expression studies, it appears that the putative
aldehyde dehydrogenase responsible for oxidizing 4-hydroxy-
butyraldehyde in P. tetrahydrofuranoxydans K1 is expressed
during growth on 1,4-butanediol but not during growth
on THF.
[{Cu6(H2NC(S)NP(S)(OiPr)2)6}{Cu3(H2NC(S)NP(S)(OiPr)2)3}ꢀ
4Me2CO]9 and an ionic aggregate [Cu10{PhNHC(S)NP(S)-
(OEt)2}9]ClO4,10 have been reported. Data available in the
literature demonstrate that complexes of coinage metal cations
with 1,3- and 1,5-bidentate ligands derived from sulfur- or
selenium-containing phosphines exhibit a clear propensity to
form oligo- and polynuclear assemblies with structural
features that depend on the conditions employed for their
preparation and the nature of the ligand. The investigation of
a Department of Chemistry, University of Stellenbosch,
Private Bag X1, Matieland 7602, Stellenbosch, South Africa
b Institut fur Anorganische Chemie, Universitat zu Koln, Greinstrasse 6,
D-50939 Koln, Germany. E-mail: damir.safin@ksu.ru,
¨
¨
¨
¨
axel.klein@uni-koeln.de; Fax: +49 221 4705196;
Tel: +49 221 4702913
w Electronic supplementary information (ESI) available: Further
experimental data. CCDC reference number 749575. For ESI and
crystallographic data in CIF or other electronic format see DOI:
10.1039/c0nj00314j
Scheme 1 Mono-oxygenation of THF.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010 New J. Chem., 2010, 34, 2835–2840 2835

View Article Online
In this work, we describe the complete structural and
spectroscopic characterization of the recently described
trinuclear CuI complex [Cu3L3], containing the thiourea ligand
tBuC(S)NHP(S)(OiPr)2 (HL).16 The complex was prepared by
an alternative method and characterised by multinuclear
NMR, IR, UV-vis spectroscopy and ESI-MS. Single crystals
for a XRD experiment were obtained from a THF solution
(in air), and contained the complex entity [Cu3L3]ꢀTHF-2-ol,
with co-crystallized 2-hydroxytetrahydrofuran. Motivated by
this finding, we also carried out catalytic experiments for the
aerobic oxidation of THF to THF-2-ol in the presence of
[Cu3L3].
RC(S)NHP(X)(OiPr)2 (X = O, S) ligands show absorption
bands at 380–420 nm, which were assigned to ligand-to-metal
(LMCT) transitions.19 Thus, we assign the long-wavelength
absorption band of [Cu3L3] to an LMCT transition.20
In the 31P{1H} NMR spectrum of [Cu3L3] in CDCl3, a
singlet signal appears at 50.2 ppm. The signal is in the region
characteristic for deprotonated N-thiophosphorylated thio-
amides and thioureas.21 The 1H NMR spectrum of [Cu3L3]
in CDCl3 contains a set of signals for the iPr protons: a
doublet for the CH3 protons at 1.32 ppm and a doublet of
septets for the CH protons at 4.75 ppm. The signal of the tBu
protons is found at 1.40 ppm. The NH proton signal is
observed at 6.63 ppm.
The crystal structure of the complex [Cu3L3]ꢀTHF-2-ol was
determined by single crystal X-ray diffraction. X-Ray suitable
crystals of the complex were obtained by slow evaporation of
the solvent from solutions in THF in air. Obviously, and to
our astonishment, [Cu3L3] provokes the formation of THF-2-ol
and traps it at recrystallization in THF.
Results and discussion
Reaction of the potassium salt KL with Cu(NO3)2 in aqueous
EtOH leads to the trinuclear [Cu3L3] complex (Scheme 2). The
complex obtained is a colorless crystalline powder. 1H,
31P{1H} NMR and IR spectroscopy data indicate that the
deprotonated thiourea Lꢁ coordinates in a 1,5-S,S0 fashion.
Recently, this complex was alternatively synthesized from
CuI.16 With this and other related experiments, we have also
demonstrated the equivalence of the structures of the formed
polynuclear CuI complexes, irrespective of the copper
‘‘source’’.7,9,16
Complex [Cu3L3]ꢀTHF-2-ol is a cyclic trimer (Fig. 1), in
which the Cu–S–Cu bridging bonds are formed by the sulfur
atoms of the CQS groups. The Cu3S3 cyclic backbone is in a
chair conformation and the copper atoms are in an S3
trigonal-planar environment. The distance Cu(2)ꢀ ꢀ ꢀCu(3)
2.764(4) (Table 1) is shorter than the sum of the van der
Waals radii of CuI.23 The other distances Cu(1)ꢀ ꢀ ꢀCu(2)
2.900(2) A and Cu(1)ꢀ ꢀ ꢀCu(3) 3.062(2) A indicate the lack of
any distinct Cuꢀ ꢀ ꢀCu interactions.10 This variation in Cuꢀ ꢀ ꢀCu
distances leads to an increase of the Cu(1)–S(6)–Cu(3) bond
angle relative to the Cu(1)–S(2)–Cu(2) and Cu(2)–S(4)–Cu(3)
The IR spectrum of [Cu3L3] shows a weak band at 605 cmꢁ1
,
assigned to the PQS groups. It is shifted to lower wave-
numbers relative to that in the spectrum of the parent ligand
HL (648 cmꢁ1) due to coordination to the metal cation.17 The
broad and strong band at 1563 cmꢁ1 is assigned to the
conjugated SCN group,18 proving the formation of an
S,S0-chelate. A unique band at 3271 cmꢁ1, related to the
tBuNH group, was also observed in the spectrum. No signal
was observed for the PNH group.
The UV-vis absorption spectrum of [Cu3L3] was recorded in
CH2Cl2 at 25 1C and is characterized by an absorption band at
348 nm (e = 4097 dm3 molꢁ1 cmꢁ1). Absorption bands for the
free ligands RC(S)NHP(X)(OiPr)2 (X = O, S) are rather high
abs
in energy (lmax o 250 nm) and thus fall into the region of
intraligand transitions.16,19 However, it is more reasonable to
compare the absorption spectrum of [Cu3L3] and the corre-
sponding potassium salt KL, and not the spectrum of the
parent ligand HL. Under the same measurement conditions,
the spectrum of KL shows a weak absorption band at 298 nm
(e = 186 dm3 molꢁ1 cmꢁ1). The band can be attributed to
intraligand transitions in the deprotonated form Lꢁ. Recently,
it was established that complexes of NiII, ZnII and CdII with
Fig.
1
A
thermal ellipsoid representation of [Cu3L3]ꢀTHF-2-ol
(THF-2-ol and hydrogen atoms were omitted for clarity). Ellipsoids
are drawn at the 50% probability level.
Table
1 Selected bond lengths [A] and bond angles [1] for
[Cu3L3]ꢀTHF-2-ol
Bond lengths
Cu(1)ꢀ ꢀ ꢀCu(2)
Cu(1)ꢀ ꢀ ꢀCu(3)
Bond angles
2.900(2)
3.062(2)
Cu(2)ꢀ ꢀ ꢀCu(3)
2.764(2)
S(1)–Cu(1)–S(2)
S(1)–Cu(1)–S(6)
S(2)–Cu(1)–S(6)
S(2)–Cu(2)–S(3)
S(2)–Cu(2)–S(4)
114.25(5)
120.89(5)
124.58(5)
119.96(5)
123.94(5)
S(3)–Cu(2)–S(4)
S(4)–Cu(3)–S(5)
S(4)–Cu(3)–S(6)
S(5)–Cu(3)–S(6)
116.08(5)
120.45(5)
123.35(5)
115.91(5)
Scheme 2 Preparation of [Cu3L3].
c
2836 New J. Chem., 2010, 34, 2835–2840 This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010

View Article Online
Chart 2 The structures of [Cu(b-diketiminate)(THF)(O2)]26 and the
proposed [Cu(THF)(O2)L].
intriguing end-on bound dioxygen (Chart 2). This complex is
considered to be a key species in the observed oxygenation
of THF.26b Therefore, it is reasonable to assume the forma-
tion of comparable complex species [Cu(THF)L] and
[Cu(THF)(O2)L]in our case (Chart 2). Due to the smaller
steric strain of our system, two THF molecules might also
coordinate to form [Cu(THF)2L]. The species detected at dP =
54.6 ppm in the 31P{1H} NMR spectrum of [Cu3L3] in THF is
thus assigned to one of the three above-mentioned species,
while the signal at 49.1 ppm corresponds to [Cu3L3].
Chart 1 A simplified line diagram of hydrogen bonding in the
molecular structure of [Cu3L3]ꢀTHF-2-ol.
Table
[Cu3L3]ꢀTHF-2-ola
2 Hydrogen bond lengths [A] and angles [1] for
D–Hꢀ ꢀ ꢀA
d(D–H)
d(Hꢀ ꢀ ꢀA)
d(Dꢀ ꢀ ꢀA)
+(DHA)
N(4)–H(25)ꢀ ꢀ ꢀS(4)0
N(6)–H(49)ꢀ ꢀ ꢀO(8)
O(8)–H(80)ꢀ ꢀ ꢀO(2)
0.88
0.88
0.84
2.68
2.24
2.14
3.500(5)
3.120(6)
2.971(6)
155
175
168
The ESI spectrum of the positive ions of complex [Cu3L3] in
THF contains the characteristic peaks for the ions [CuL]+,
[Cu2L2]+ and [Cu3L2]+. The molecular ion [Cu3L3 + H]+
(m/z = 1125.1) is stable under the measurement conditions
and was found in the mass spectrum of the complex. However,
the ion has intensity of 17%. The most intense peak corres-
ponds to the ion [Cu4L3]+ (m/z = 1187.6). Recently the CuI
a
Symmetry transformations used to generate the S(4)0 atom: (1 ꢁ x,
2 ꢁ y, ꢁz).
bond angles (Table 1). The three CuSCNPS six-membered
chelate rings are similar to that in the mononuclear analogue,
which adopts a distorted boat conformation. In these chelate
rings, the CQS and PQS bonds are lengthened, while the C–N
and P–N bonds are shortened in comparison to typical values
for N-thiophosphorylated thioureas and thioamides.21 In the
polynuclear complex [Cu3L3]ꢀTHF-2-ol, the CQS bonds,
participating by the formation of bridges, are especially
lengthened to values of 1.781(5), 1.785(5) and 1.807(5) A,
characteristic for single bonds.24
complexes [{Cu4QIII3}{CuCl2}]ꢀCCl4 and [Cu4QIII3]ꢀI3 (QIII
=
[Ph2P(S)NP(S)PPh2]ꢁ) were described.5a,g It was established
+
that the tetrahedral complex core [Cu4QIII
]
(Chart 3) is
3
rather stable, even in the presence of acids.5g It is noteworthy
that the second intense peak corresponds to the ion
[Cu(THF)(O2)L
+
H]+ [m/z (%)
=
480.1 (93)].
Furthermore, the ion [Cu(THF)L + H]+ was also assigned,
while no ions with a mass characteristic of the species
[Cu(THF)2L] were determined.
[Cu3L3]ꢀTHF-2-ol contains, in the crystal, two intermolecular
N–Hꢀ ꢀ ꢀSQC hydrogen bonds between two neighboring
molecules (Chart 1, Table 2). As a result of the intermolecular
interactions, a dimer is formed. There are also intermolecular
O–Hꢀ ꢀ ꢀO–iPr and N–Hꢀ ꢀ ꢀO–H hydrogen bonds between the
hydroxy group of co-crystallized THF-2-ol, and the NH and
OiPr groups of the trinuclear molecule [Cu3L3]ꢀTHF-2-ol
(Chart 1). To the best of our knowledge, the complex
[Cu3L3]ꢀTHF-2-ol is the second example, showing the crystal
structure of THF-2-ol. A polynuclear CoIII complex with
tBuCOOꢁ, containing solvated THF-2-ol, has been
published.25 Although the preparation method of THF-2-ol
was described, no discussion of a mechanism of formation was
given by the authors.
The ESI spectrum of the complex [Cu3L3] using negative
ionisation contains characteristic peaks for the ions [L]ꢁ and
[CuL2]ꢁ. The [CuL2]ꢁ backbone was previously found by us in
the polynuclear CuI complex [KCuQIV
]
2 n
(QIV
=
[PhC(S)NP(S)P(OiPr)2]ꢁ).22 The most intense peak observed
in the spectrum corresponds to the ion [Cu2L3 + THF]ꢁ
(m/z = 1132.7).
Trying to find further evidence for the dioxygen-containing
complex [Cu(THF)(O2)L], we carried out resonance Raman
spectroscopy (rR) in the region 500–1500 cmꢁ1. We found a
number of bands corresponding to Lꢁ but no band
Attempts to record NMR spectra of the trapped THF-2-ol
in [Cu3L3]ꢀTHF-2-ol unfortunately failed. However, when
dissolving bulk [Cu3L3] in THF-d8, the 31P{1H} NMR
spectrum revealed two singlet signals at 49.1 and 54.6 ppm
with relative intensities of 7.3 : 1 for the high-field signal.
Recently it was reported that the CuI complex fragment
[Cu(b-diketiminate)] (depicted in Chart 2) is able to bind
dioxygen in a side-on fashion.26 At the same time, nitriles
RCN and THF can be bound either alternatively to O2
or at the same time as, for example, in the calculated species
[Cu(b-diketiminate)(THF)(O2)]. This complex exhibits an
Chart 3 The structure of [Cu4QIII + (QIII = [Ph2P(S)NP(S)PPh2]ꢁ).5a,g
]
3
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010 New J. Chem., 2010, 34, 2835–2840 2837

View Article Online
unambiguously attributable to the dioxygen ligand, which
might exist in a neutral, anionic (superoxide) or dianionic
(peroxide) form, was detected. Thus, ESI mass spectrometry is
not only a promising technique for the structure determination
irradiation in the presence of the copper complex is practically
negligible (Table S1w). Summing up, all the products lead to
conversions of about 30–35%, which corresponds to approxi-
mately 0.01 mmol of the starting THF amount. Thus, no
unequivocal proof for a catalytic reaction is supplied by our
experiments. After 2 d stirring the reaction mixture under an
ambient atmosphere, the formation of THF-2-ol is practically
stopped; we suppose that complex decomposition is the main
reason. Solutions containing [Cu3L3] became greenish-blue
and we assumed that CuII compounds were formed. In an
experiment to elucidate the underlying species, such a solution
was left standing for 3 d, after which the solvent was removed
by vacuum. The residue was extracted using toluene and
analyzed by EPR spectroscopy. It was established that the
EPR spectra were similar to the spectra of the previously
described CuII complexes [Cu{RNHC(S)NP(O)(OiPr)2-N,S}2]
(R = Ph, c-hex).27
of polynuclear CuI complexes with RC(S)NHP(S)R0 ligands,
2
but in the present case has also allowed us to establish the
formation of [Cu(THF)L] and [Cu(THF)(O2)L], and to
exclude the formation of [Cu(THF)2L].
Unfortunately, we do not have further evidence for the
assumed species [Cu(THF)(O2)L] at the moment, and
therefore we cannot tell anything about the oxidation state
of the copper or the dioxygen ligand. Tolman et al. showed
that their complexes exhibited significant CuIII-peroxo
character.26 Efforts to clarify the situation for our system by
vibrational spectroscopy and XAS will be made in the near
future.
Inspired by the crystallization of [Cu3L3]ꢀTHF-2-ol, we have
studied the formation of THF-2-ol in THF under different
conditions (Table S1 in the ESIw). In a typical procedure,
complex [Cu3L3] was dissolved in freshly distilled and
de-gassed THF, and the samples were further analyzed by
GC-MS. The following observations were made (Table S1w).
In the absence of O2 (under argon), very small amounts of
THF-2-ol and 4-butyrolactone were always found (1–2%),
which were probably due to traces of O2 in the THF solution.
The absence or presence of [Cu3L3] had no impact on this. The
presence of [Cu3L3] accelerated the formation of THF-2-ol,
while the 4-butyrolactone concentration remained largely the
same and generally small in all the experiments. After 24 h,
about 15% of THF-2-ol and after 60 h about 25% of THF-2-ol
were found. After an initiation period of about 6–10 h,
appreciable amounts of 2,3-dihydrofuran were observed and
reached a maximum concentration of about 7% after 60 h.
We assume that the latter is the product of dehydratation of
THF-2-ol. Under this assumption, we may add this 7% to the
25% of THF-2-ol formed after 60 h. Furthermore, according
to the obtained data, we may assert that the influence of UV
Furthermore, we examined complexes [Cu3QI3]7 and
[(Cu3QII3)2]8 to see if they also provoked the aerobic oxidation
of THF to THF-2-ol. Surprisingly, both complexes were
extremely unstable under the experimental conditions
(bubbling O2 through a catalyst solution in THF). Solutions
containing the catalyst [Cu3QI3] or [(Cu3QII3)2] also became
greenish-blue after a few minutes of bubbling due to the
formation of CuII species, which were detected by EPR
spectroscopy.27 Thus, we postulate that the starting
complex [Cu3L3] decomposes with the formation of the CuII
complex [Cu{tBuNHC(S)NP(O)(OiPr)2-N,S}2], which is
unstable in solutions and further decomposes. A signal for
(iPrO)2P(O)OH at 2–3.5 ppm (dP = 2.98 ppm)28 was observed
in 31P{1H} NMR spectra of reaction mixtures, though
complex [Cu3L3] could be isolated quantitatively in a shorter
time, e.g. 1.5 d (yield 98%). We also examined the formation
of THF-2-ol in the vice versa procedure: complex [Cu3L3] was
treated with oxygen (bubbling O2 through the solid material)
for 1 d and then THF was added. First, it is noteworthy that
the solid material [Cu3L3] did not decompose under the
Scheme 3 The supposed catalytic mechanism of THF-2-ol formation in the presence of [Cu3L3].
c
2838 New J. Chem., 2010, 34, 2835–2840 This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010

View Article Online
influence of oxygen, and secondly the obtained yields of
THF-2-ol were very similar to those listed in the entry 7 of
Table S1.w Furthermore, using morpholine or 1,4-dioxane
instead of THF as a substrate did not lead to the corresponding
oxygenation products. Further investigation of the formation
of THF-2-ol under an 18O2 atmosphere (using GC-MS)
proved dioxygen activation.
liquid N2-cooled CCD detector. EPR measurements were
performed on Bruker ESP 300 spectrometer equipped with
X-band 4110ST resonator and standard nitrogen flow acces-
sory. The field was controlled by Bruker ER035 field meter
and the microwave frequency measured with Hewlett–Packard
5255A frequency counter. Electrospray ionization mass
spectra were measured using a Finnigan-Mat TCQ 700 mass
spectrometer on 10ꢁ6 M solutions in THF. The speed of
sample submission was 2 mL minꢁ1. The ionization energy
was 4.5 kV. The capillary temperature was 200 1C. The
GC-MS analysis was carried out using an Automass
150 instrument (Delsi Nermag, France) equipped with a
0.25 mm ꢂ 25 m capillary column with 0.3 mm OV101 (helium
as carrier gas, 0.6 atm; injector temperature 250 1C, divider
1 : 40, volume of the sample 0.2 mL, initial column temperature
50 1C (2 min), temperature programming to 250 1C at a rate of
10 1C minꢁ1, interface temperature 250 1C). UV irradiation
experiments were performed using a CAMAG UV Lamp.
Provided that the formation of the complex
[Cu(THF)(O2)L] is one of the key steps for the oxygenation
of THF, we can speculate that the [Cu(THF)(O2)L] complex
(Chart 1) is an intermediate in our system, while the
hypothetical complexes [Cu(THF)(O2)QI] and [Cu(THF)(O2)QII]
were not formed or were too unstable. The combination of the
bulky tBuNH and iPrO substituents might be crucial in
stabilizing the complex [Cu(THF)(O2)L], while the less bulky
systems [Cu3QI3] and [(Cu3QII3)2] rapidly decompose.
At present, we cannot provide an exact mechanism for the
oxygenation of THF with [Cu3L3], but the virtual absence of
4-butyrolactone supposes that the reaction follows other
mechanisms than those described in the biological systems
mentioned above. The assumed catalytic mechanism is shown
in Scheme 3. We suppose that the key role is played by
H-bonded structure A (Scheme 3). It is noteworthy that
the existence of the similar hydrogen bonded structures
[Cu(b-diketiminate)(X)(O2)], where X is carbene, pyrazole,
tetrazole and 2-pyridinamine, was described previously.26b
Syntheses
tBuNHC(S)NHP(S)(OiPr)2 (HL) was synthesized as described
previously.17
[Cu3(tBuNHC(S)NP(S)(OiPr)2-S,S0)3] ([Cu3L3]). A suspen-
sion of HL (0.936 g, 3 mmol) in aqueous ethanol (25 mL) was
mixed with an ethanol solution of KOH (0.185 g, 3.3 mmol).
The resulting mixture was added dropwise to a suspension of
Cu(NO3)2ꢀ6H2O (0.474 g, 1.6 mmol) in aqueous ethanol
(20 mL). The mixture was stirred at room temperature for
5 h and left overnight. The resulting complex was extracted
using CH2Cl2, and the combined extracts were dried over
anhydrous MgSO4. The solvent was then removed in vacuum
and the residue was recrystallized from THF. The complex
was obtained as a colorless microcrystalline material. Yield:
Conclusions
In summary, the reaction of Cu(NO3)2 with the potassium salt
of HL has allowed us to obtain the trinuclear complex [Cu3L3].
NMR and IR experiments have shown that the deprotonated
thiourea Lꢁ acts as a 1,5-S,S0-ligand. Single crystals obtained
from THF solutions grown under aerial conditions revealed
the complex [Cu3L3]ꢀTHF-2-ol, which is a cyclic trimer,
formed by bridging bonds with participation of the thio-
carbonyl sulfur atoms, and two molecules are linked through
two hydrogen bonds of the N–Hꢀ ꢀ ꢀSQC type. The metal
cations are in a trigonal planar S3 environment. Surprisingly,
the compound contains one equivalent of 2-hydroxytetra-
hydrofuran (THF-2-ol). Further experiments (NMR, ESI-MS)
and catalytic studies established that [Cu3L3] provokes
the formation of 2-hydroxotetrahydrofuran and traps it by
recrystallization from THF.
0.821 g, 73%. M.p. 139 1C. 1H NMR (CDCl3): 1.32 (d, 3JH,H
6.1 Hz, 36 H, CH3, iPr), 1.40 (s, 27 H, CH3, tBu), 4.75 (d. sept,
=
3JH,H = 6.0 Hz, 3JP,H = 10.5 Hz, 6 H, OCH), 6.63 (d, 4JP,H
=
9.2 Hz, 3 H, NH) ppm. 31P{1H} NMR (CDCl3): 50.2 ppm.
31P{1H} NMR (THF): 49.1 (7.3 P), 54.6 (1 P) ppm. IR:
605 (PQS), 1000, 1009 (POC), 1563 (SCN), 3271 (NH) cmꢁ1
.
ESI-MS (positive ion): m/z (%) = 376.1 (11) [CuL + H]+,
447.8 (18) [Cu(THF)L + H]+, 480.1 (93) [Cu(THF)(O2)L + H]+,
519.4 (3) [Cu(THF)2L + H]+, 749.5 (21) [Cu2L2 + H]+, 885.2
(29) [Cu3L2 + THF]+, 1125.1 (17) [Cu3L3 + H]+, 1187.6
(100) [Cu4L3]+. ESI-MS (negative ion): m/z (%) = 311.2 (39)
[L]ꢁ, 686.3 (22) [CuL2]ꢁ, 1132.7 (100) [Cu2L3 + THF]ꢁ. Anal.
calc. for C33H72Cu3N6O6P3S6 (1124.91): C, 35.23; H, 6.45, N,
7.47. Found: C, 35.41; H, 6.39; N, 7.51.
Experimental
General procedures
Catalytic properties of [Cu3L3] in the oxidation of THF
Elemental analysis was performed on a HEKAtech CHNS
EuroEA 3000 microanalyser. NMR spectra (CDCl3 or THF-d8)
were obtained on a Bruker Avance 300 MHz spectrometer at
25 1C. Infrared spectra (Nujol) were recorded with a Specord
M-80 spectrometer in the range 400–3600 cmꢁ1. Electronic
absorption spectra of 10ꢁ4 M solution in CH2Cl2 were
In a typical procedure, the complex [Cu3L3] (0.01125 g,
0.01 mmol) was dissolved in freshly distilled and de-gassed
THF (3 mL) at room temperature under an argon atmosphere
in the dark or under UV irradiation. Further experimental
conditions are summarized in Table S1 in the ESI.w
measured on
a Lambda-35 spectrometer in the range
X-Ray crystallography
200–1000 nm. Resonance Raman spectra were collected on
an Acton AM-506 spectrometer using Kaiser Optical
holographic super-notch filters with a Princeton Instruments
Data were collected on a Bruker SMART Apex CCD
diffractometer with graphite monochromated Mo-Ka
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010 New J. Chem., 2010, 34, 2835–2840 2839

View Article Online
radiation (l = 0.71073 A).29 Data reduction was carried
out using SAINT.30 Empirical absorption correction was
performed with SADABS.31 The structure was solved by
direct methods, while the remainder of the atomic positions
were found using difference Fourier methods. All non-hydrogen
atoms were refined anisotropically (with appropriate restraints
using SIMU and ISOR) by blocked full-matrix least squares
calculations on F2 using SHELX-9732 within the X-seed
environment. Hydrogen atoms were added to the structure
model in calculated positions and were refined as rigid atoms.
ORTEP-III for Windows was used to generate the figure at
the 50% probability level.33 C33H72Cu3N6O6P3S6, C4H8O2,
M = 1213.05, monoclinic, space group P21/n, a = 14.448(7),
b = 15.407(7), c = 25.575(11) A, b = 91.100(8)1, V = 5692(5) A3,
T = 100(2) K, Z = 4, m(Mo-Ka) = 1.462 mmꢁ1, F(000) =
9 R. C. Luckay, X. Sheng, C. E. Strasser, H. G. Raubenheimer,
D. A. Safin, M. G. Babashkina and A. Klein, Dalton Trans., 2009,
4646.
10 M. L. Niven, P. Kyriacou and T. A. Modro, J. Chem. Soc., Dalton
Trans., 1988, 1915.
11 C. Bock, R. M. Kroppenstedt and H. Diekmann, Appl. Microbiol.
Biotechnol., 1996, 45, 408.
12 (a) R. E. Parales, J. E. Adamus, N. White and H. D. May, Appl.
Environ. Microbiol., 1994, 60, 4527; (b) S. Mahendra and
L. Alvarez-Cohen, Int. J. Syst. Evol. Microbiol., 2005, 55, 593.
13 (a) U. Kohlweyer, B. Thiemer, T. Schrader and J. R. Andreesen,
FEMS Microbiol. Lett., 2000, 186, 301; (b) P. Kaempfer,
U. Kohlweyer, B. Thiemer and J. R. Andreesen, Int. J. Syst. Evol.
Microbiol., 2006, 56, 1535.
14 K. J. Daye, J. C. Groff, A. C. Kirpekar and R. Mazumder, J. Ind.
Microbiol. Biotechnol., 2003, 30, 705.
15 S. Vainberg, K. McClay, H. Masuda, D. Root, C. Condee,
G. J. Zylstra and R. J. Steffan, Appl. Environ. Microbiol., 2006,
72, 5218.
16 M. G. Babashkina, Chelate complexes of N-thioacylamidothio-
phosphates with copper(I) cation, PhD thesis, Kazan State Uni-
versity, Kazan, Russia, 2006, p. 166.
2544, 32467 reflections measured, 11760 observed (Rint
=
0.067). R indices (all data): R1 = 0.0696, wR2 = 0.1550.w
17 D. A. Safin, F. D. Sokolov, T. R. Gimadiev, V. V. Brusko,
M. G. Babashkina, D. R. Chubukaeva, D. B. Krivolapov and
I. A. Litvinov, Z. Anorg. Allg. Chem., 2008, 634, 967.
18 E. G. Yarkova, N. R. Safiullina, I. G. Chistyakova, N. G. Zabirov,
F. M. Shamsevaleev and R. A. Cherkasov, Zh. Obshch. Khim.
(USSR), 1990, 60, 1790.
References
1 (a) P. C. Ford, E. Cariati and J. Bourassa, Chem. Rev., 1999, 99,
3625; (b) C.-M. Che and S.-W. Lai, Coord. Chem. Rev., 2005, 249,
1296; (c) D. L. Phillips, C.-M. Che, K. H. Leung, Z. Mao and
M.-C. Tse, Coord. Chem. Rev., 2005, 249, 1476; (d) K. Saito,
T. Arai, N. Takahashi, T. Tsukuda and T. Tsubomura, Dalton
Trans., 2006, 4444.
2 A. Zazybin, O. Osipova, U. Khusnutdinova, I. Aristov,
B. Solomonov, F. Sokolov, M. Babashkina and N. Zabirov,
J. Mol. Catal. A: Chem., 2006, 253, 234.
19 (a) V. V. Brusko, A. I. Rakhmatullin, V. G. Shtyrlin,
D. B. Krivolapov, I. A. Litvinov and N. G. Zabirov, Polyhedron,
2006, 25, 1433; (b) D. A. Safin, F. D. Sokolov, H. Noth,
¨
M. G. Babashkina, T. R. Gimadiev, J. Galezowska and
H. Kozlowski, Polyhedron, 2008, 27, 2022; (c) D. A. Safin,
A. Klein, M. G. Babashkina, H. Noth, D. B. Krivolapov,
¨
I. A. Litvinov and H. Kozlowski, Polyhedron, 2009, 28, 1504.
20 C. Kutal, Coord. Chem. Rev., 1990, 99, 213.
3 (a) M. Afzaal, D. J. Crouch, P. O’Brien, J. Raftery, P. J. Skabara,
A. J. P. White and D. J. Williams, J. Mater. Chem., 2004, 14, 233;
21 F. D. Sokolov, V. V. Brusko, N. G. Zabirov and R. A. Cherkasov,
Curr. Org. Chem., 2006, 10, 27.
22 M. G. Babashkina, A. I. Rakhmatullin, D. A. Safin, F. Fayon,
F. D. Sokolov, A. Klein, D. B. Krivolapov, T. Pape, F. E. Hahn
and H. Kozlowski, Dalton Trans., submitted.
23 J. E. Huheey, E. A. Keiter and R. L. Keiter, in Inorganic
Chemistry: Principles of Structure and Reactivity, Harper Collins
College Publishers, New York, 4th edn, 1993.
24 V. N. Solov’ev, N. G. Zabirov, R. A. Cherkasov and
I. V. Martynov, Zh. Strukt. Khim. (USSR), 1991, 32, 80.
25 (a) T. O. Denisova, M. A. Golubnichaya and S. E. Nefedov, Russ.
Chem. Bull., 2003, 52, 2760; (b) S. E. Nefedov and T. O. Denisova,
Russ. J. Inorg. Chem., 2006, 51, 1404.
26 (a) N. W. Aboelella, S. V. Kryatov, B. F. Gherman,
W. W. Brennessel, V. G. Young, Jr., R. Sarangi, E. V. Rybak-
Akimova, K. O. Hodgson, B. Hedman, E. I. Solomon,
C. J. Cramer and W. B. Tolman, J. Am. Chem. Soc., 2004, 126,
16896; (b) D. E. Heppner, B. F. Gherman, W. B. Tolman and
C. J. Cramer, Dalton Trans., 2006, 4773.
27 R. R. Garipov, V. G. Shtyrlin, D. A. Safin, Y. I. Zyavkina,
F. D. Sokolov, A. L. Konkin, A. V. Aganov and
A. V. Zakharov, Chem. Phys., 2006, 320, 59.
28 R. R. Abdreimova, D. N. Akbayeva, G. S. Polimbetova,
A.-M. Caminade and J.-P. Majoral, Phosphorus, Sulfur Silicon
Relat. Elem., 2000, 156, 239.
29 SMART Data Collection Software, Version 5.629, Bruker AXS
Inc., Madison, WI, 2003.
(b) A. Grodzicki, I. Łakomska, P. Piszczek, I. Szymanska and
´
E. Sz"yk, Coord. Chem. Rev., 2005, 249, 2232.
4 G. Henkel and B. Krebs, Chem. Rev., 2004, 104, 801.
5 (a) O. Siiman, C. P. Huber and M. L. Post, Inorg. Chim. Acta,
1977, 25, L11; (b) D. J. Birdsall, A. M. Z. Slawin and
D. J. Woollins, Inorg. Chem., 1999, 38, 4152; (c) S. Canales,
O. Crespo, M. Concepcion Gimeno, P. G. Jones, A. Laguna,
A. Silvestru and C. Silvestru, Inorg. Chim. Acta, 2003, 347, 16;
(d) H. Liu, N. A. G. Bandeira, M. J. Calhorda, M. G. B. Drew,
V. Felix, J. Novosad, F. Fabrizi de Biani and P. Zanello,
J. Organomet. Chem., 2004, 689, 2808; (e) D. Rusanova,
W. Forsling, O. N. Antzutkin, K. J. Pike and R. Dupree, Lang-
muir, 2005, 21, 4420; (f) P. Moore, W. Errington and S. P. Sangha,
Helv. Chim. Acta, 2005, 88, 782; (g) M. C. Aragoni, M. Arca,
M. B. Carrea, F. Demartin, F. A. Devillanova, A. Garau,
M. B. Hursthouse, S. L. Huth, F. Isaia, V. Lippolis,
H. R. Ogilvie and G. Verani, Eur. J. Inorg. Chem., 2006, 200;
(h) C. W. Liu, I.-J. Shang, R.-J. Fu, B.-J. Liaw, J.-C. Wang and
I.-J. Chang, Inorg. Chem., 2006, 45, 2335; (i) Y.-J. Hsu,
C.-M. Hung, Y.-F. Lin, B.-J. Liaw, T. S. Lobana, S.-Y. Lu and
C. W. Liu, Chem. Mater., 2006, 18, 3323; (j) D. Rusanova,
W. Forsling, O. N. Antzutkin, K. J. Pike and R. Dupree,
J. Magn. Reson., 2006, 179, 140; (k) W. Shi, M. Shafaei-Fallah,
C. E. Anson and A. Rothenberger, Dalton Trans., 2006, 3257.
6 E. Herrmann, R. Richter and N. T. T. Chau, Z. Anorg. Allg.
Chem., 1997, 623, 403.
7 F. D. Sokolov, M. G. Babashkina, D. A. Safin,
A. I. Rakhmatullin, F. Fayon, N. G. Zabirov, M. Bolte,
V. V. Brusko, J. Galezowska and H. Kozlowski, Dalton Trans.,
2007, 4693.
8 F. D. Sokolov, M. G. Babashkina, F. Fayon, A. I. Rakhmatullin,
D. A. Safin, T. Pape and F. E. Hahn, J. Organomet. Chem., 2009,
694, 167.
30 SAINT, Data Reduction Software, Version 6.45, Bruker AXS Inc.,
Madison, WI, 2003.
31 SADABS, Version 2.05, Bruker AXS Inc., Madison, WI, 2002.
32 G. M. Sheldrick, SHELX-97, Program for Crystal Structure
Analysis, University of Gottingen, Germany, 1997.
¨
33 L. J. Farrugia, J. Appl. Crystallogr., 1997, 30, 565.
c
2840 New J. Chem., 2010, 34, 2835–2840 This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2010